{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:ZPESEEWXE6CXIU53C4RGKE34G4","short_pith_number":"pith:ZPESEEWX","schema_version":"1.0","canonical_sha256":"cbc92212d727857453bb172265137c3711d0a31ccd48076cd3bd18d40917dc47","source":{"kind":"arxiv","id":"2401.08406","version":3},"attestation_state":"computed","paper":{"title":"RAG vs Fine-tuning: Pipelines, Tradeoffs, and a Case Study on Agriculture","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Angels Balaguer, Bruno Silva, Daniel Holstein, Jennifer Marsman, Leonardo O. Nunes, Morris Sharp, Nick Mecklenburg, Rafael Padilha, Ranveer Chandra, Renato Luiz de Freitas Cunha, Roberto de M. Estev\\~ao Filho, Sara Malvar, Swati Sharma, Todd Hendry, Vijay Aski, Vinamra Benara","submitted_at":"2024-01-16T14:44:47Z","abstract_excerpt":"There are two common ways in which developers are incorporating proprietary and domain-specific data when building applications of Large Language Models (LLMs): Retrieval-Augmented Generation (RAG) and Fine-Tuning. RAG augments the prompt with the external data, while fine-Tuning incorporates the additional knowledge into the model itself. However, the pros and cons of both approaches are not well understood. In this paper, we propose a pipeline for fine-tuning and RAG, and present the tradeoffs of both for multiple popular LLMs, including Llama2-13B, GPT-3.5, and GPT-4. Our pipeline consists "},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2401.08406","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.CL","submitted_at":"2024-01-16T14:44:47Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"108e10b3209299e325c87a75c3d86aa8b7e35b96b92244905a0bb098dec44367","abstract_canon_sha256":"960ebbf36ac171b15245ac4d8c7d51d2b5741af4c46b1d1fa3be55914ddd30dc"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:39:12.766861Z","signature_b64":"T31JMoApwSjciRjHeqkM9bR2a8cteJOZ9p+Qba9odlGCMJOGwY2UcSKh+aS1aesaBwqb17ZJAWudChaJqA05Dg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"cbc92212d727857453bb172265137c3711d0a31ccd48076cd3bd18d40917dc47","last_reissued_at":"2026-07-05T07:39:12.766332Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:39:12.766332Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"RAG vs Fine-tuning: Pipelines, Tradeoffs, and a Case Study on Agriculture","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Angels Balaguer, Bruno Silva, Daniel Holstein, Jennifer Marsman, Leonardo O. Nunes, Morris Sharp, Nick Mecklenburg, Rafael Padilha, Ranveer Chandra, Renato Luiz de Freitas Cunha, Roberto de M. Estev\\~ao Filho, Sara Malvar, Swati Sharma, Todd Hendry, Vijay Aski, Vinamra Benara","submitted_at":"2024-01-16T14:44:47Z","abstract_excerpt":"There are two common ways in which developers are incorporating proprietary and domain-specific data when building applications of Large Language Models (LLMs): Retrieval-Augmented Generation (RAG) and Fine-Tuning. RAG augments the prompt with the external data, while fine-Tuning incorporates the additional knowledge into the model itself. However, the pros and cons of both approaches are not well understood. In this paper, we propose a pipeline for fine-tuning and RAG, and present the tradeoffs of both for multiple popular LLMs, including Llama2-13B, GPT-3.5, and GPT-4. Our pipeline consists "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2401.08406","kind":"arxiv","version":3},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2401.08406/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"2401.08406","created_at":"2026-07-05T07:39:12.766400+00:00"},{"alias_kind":"arxiv_version","alias_value":"2401.08406v3","created_at":"2026-07-05T07:39:12.766400+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2401.08406","created_at":"2026-07-05T07:39:12.766400+00:00"},{"alias_kind":"pith_short_12","alias_value":"ZPESEEWXE6CX","created_at":"2026-07-05T07:39:12.766400+00:00"},{"alias_kind":"pith_short_16","alias_value":"ZPESEEWXE6CXIU53","created_at":"2026-07-05T07:39:12.766400+00:00"},{"alias_kind":"pith_short_8","alias_value":"ZPESEEWX","created_at":"2026-07-05T07:39:12.766400+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":7,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.23193","citing_title":"CultivAgents: Cultivating Relationship-Centered Multi-Agent Systems for Personalized Gardening","ref_index":2,"is_internal_anchor":false},{"citing_arxiv_id":"2505.01307","citing_title":"Document Retrieval Augmented Fine-Tuning (DRAFT) for safety-critical software assessments","ref_index":2,"is_internal_anchor":false},{"citing_arxiv_id":"2601.06352","citing_title":"CARD: Cluster-level Adaptation with Reward-guided Decoding for Personalized Text Generation","ref_index":1,"is_internal_anchor":false},{"citing_arxiv_id":"2406.00515","citing_title":"A Survey on Large Language Models for Code Generation","ref_index":90,"is_internal_anchor":false},{"citing_arxiv_id":"2311.05232","citing_title":"A Survey on Hallucination in Large Language Models: Principles, Taxonomy, Challenges, and Open Questions","ref_index":73,"is_internal_anchor":false},{"citing_arxiv_id":"2605.10640","citing_title":"Towards Understanding Continual Factual Knowledge Acquisition of Language Models: From Theory to Algorithm","ref_index":48,"is_internal_anchor":false},{"citing_arxiv_id":"2605.09533","citing_title":"Assessment of RAG and Fine-Tuning for Industrial Question-Answering-Applications","ref_index":24,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/ZPESEEWXE6CXIU53C4RGKE34G4","json":"https://pith.science/pith/ZPESEEWXE6CXIU53C4RGKE34G4.json","graph_json":"https://pith.science/api/pith-number/ZPESEEWXE6CXIU53C4RGKE34G4/graph.json","events_json":"https://pith.science/api/pith-number/ZPESEEWXE6CXIU53C4RGKE34G4/events.json","paper":"https://pith.science/paper/ZPESEEWX"},"agent_actions":{"view_html":"https://pith.science/pith/ZPESEEWXE6CXIU53C4RGKE34G4","download_json":"https://pith.science/pith/ZPESEEWXE6CXIU53C4RGKE34G4.json","view_paper":"https://pith.science/paper/ZPESEEWX","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2401.08406&json=true","fetch_graph":"https://pith.science/api/pith-number/ZPESEEWXE6CXIU53C4RGKE34G4/graph.json","fetch_events":"https://pith.science/api/pith-number/ZPESEEWXE6CXIU53C4RGKE34G4/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ZPESEEWXE6CXIU53C4RGKE34G4/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ZPESEEWXE6CXIU53C4RGKE34G4/action/storage_attestation","attest_author":"https://pith.science/pith/ZPESEEWXE6CXIU53C4RGKE34G4/action/author_attestation","sign_citation":"https://pith.science/pith/ZPESEEWXE6CXIU53C4RGKE34G4/action/citation_signature","submit_replication":"https://pith.science/pith/ZPESEEWXE6CXIU53C4RGKE34G4/action/replication_record"}},"created_at":"2026-07-05T07:39:12.766400+00:00","updated_at":"2026-07-05T07:39:12.766400+00:00"}